- Email: [email protected]
Dissolved aluminium in interstitial waters of recent marine sediments
NOTE
Dissolved
aluminium in interstitial marine sediments SKRG~ CASCH~:TKJ and
Laboratoire
d’Oc&nographie.
Universitti
Librc
waters of recent
ROLANI) WOLLAST
de Bruxelles.
A\. F. Roosevelt.
50. Bruxelles
1050. Belgrum
Abstract-Measurements of the concentration and vertical distrlhution of dissolved aluminium and silica. and of pH. in interstitial waters of recent marine sediments from the North Sea and the Meditcrranean Sea were performed to evaluate the behaviour of aluminium during early diagenesis. The results suggest that thermodynamic equilibria alone do not control the concentratrons of dissolved species in the system AI-Si-O,-H,O during early diagenesis. Rather. these concentrations are governed by dynamic factors involving mineral dissolution-precipitation reactions and diffusion.
INTRODUCTION Knowledge of the concentration and distribution of dissolved aluminium (Al,,,,) in natural waters has improved substantialI> since the development of Al fluorimetric method. In the case of seawater. the vertical distribution of Al,, in the water column strongly suggests the importance of the planktonic activity on the removal of AI from the oceanic system (MA~KI:NZIL CI al.. 1978; CASCHETTO and WOLLAST. to be published). The behaviour of Al, in estuarine environments is still a matter of controversy (HBOKAWA (‘I ul.. 1970: HYUES and LISS. 1977). To our knowledge. there are no data available on the distribution of Al,., in the interstitial waters of marine sedlmen&. However. to understand the geochemical cycle of Al and processes involved in the earl) diagenesis of cla! minerals. it is necessarv to know the concentration. distribution. and mineral sinks and sources of Al,,., in marine sediments. This brief contribution provides some preltmmarq results from coastal marine sediments concerning these factors. METHODOLOCI Gene-al
Five cores were collected. two from the Mediterranean Sea and three from the North Sea. The Mediterranean cores 57-2 and 94 were obtained using a gravity corer about 2.7 km offshore of Calvi (Corsica) at depths of 57 and 94 m. The core sediments consist principally of medium-grained. quartzo-feldspathic sand containing accessor! biotite. calcareous tests. siliceous sponge spicules and numerous remains of marine phanerogams. Cores 19 and 32 were taken in the Fladden Ground of the North Sea (58.25’5”N. 00’02’O”E and 58’50’3”N. 00’41’4”E) at depths of 135 and 140m by means of a box-corer during cruise 54-6 of the K~~orr (“FLEX” program). These sediments are medium-grained sands. Glacial pebbles are recognizable in core 33. The third North Sea core was collected b) Belgian Nav) divers about 1.8 km offshore of the Belgian Coast at a depth of 10m (Station Mll49). It IS characterized b> a ver) fine. dark mudd! sand. No laminations are present in the five cores.
Core 57-2 was shced Immcdlatel~ after collection on board ship. and then stored in sealed polyethylene bags for 3 hr at IO C. The interstitial waters were extracted from the sediments using a PVC ‘squeezer’ by pressure filtration through a 0.45 blrn Milhpore” filter under 3 bar of nitrogen. Cores 19. 32.94. and M 1 149 were frozen immediately after collection. The interstitial waters of cores 19. 32 and 94 were later extracted following the above procedure. Waters from core Ml 149 however. were removed from the sediments by centrifugation and subsequent filtration. In all cases. the pore waters were extracted a~ the i/l .sirrr temperatures of collection. Frozen cores were thawed overnight at the iu .\itu temperature before treatment.
Interstitial waters were analyzed for dissolved silica b) absorption spectrophotometrq (Technicon Auto-Analyzer. Methodology 186-72W) and for dissolved aluminium by fluorescence spectrophotometry of the Al-lumogallion complex (NISHIKAWA (0 ul.. 1967: SHIGKMATSI. or al.. 1970: HYI>ES and LISS. 1976). using the method of Al increments. In all the pore waters we analyzed. this Al incremental calibration was able to prevent the problems involving interferences owing to F. SOi-. Fe and dissolved organic matter as previously shown by Hsuts and Ltss (1976). However. if sediments contain much organic matter. it must be photo-oxidized (IOOOW U.V. Hanovia lamp) prior to the AI,, analysts. as was done in the case of core M 1 149. The practical detectlon limit of the Al-lumogallion method was found to be 0.003~mole Al I-‘. From five independent analyses of a seawater containing 0.078 pmole Al I-‘, we obtained an estimated standard deviation of 0.003 pmole Al I_ ‘. To determine the effect of storage on Al,, determinations. a duplicate analysis of Al,, was performed on the interstitial waters of core 57-2 aker storage of the waters for IO days at 7 C (Fig. 2). After storage most of the water samples exhibited a decrease in Al,, but the general shape of the vertical concentration gradient of Al,, was maintained. HYDE and LISS 11976) investigated variations m the amount of AI,,, detected bj this fluorimetric method in claq suspensions of various concentrations and in natural waters which were filtered through different pore size mem425
Nom
426
OcDth
I
unfrozen
a
sediment
Al,,
( PMI-’ I
b
Fig. I. Al.,, concentration in interstitial waters of sediments collected m ,tn artifiaal lagoon on the Belgian Const: (al comp~t~son between the vertical distribLition in mterstittal waters squeezed from the unfroam sediment (0) and the vertical distributton in interstitial waters squeezed from frozen sediment ( + I: (b) plot of the values from unfrozen sediment tersus those from frozen sediment. The slope of the line through the points IS 0.X7. hranrs. They were able to show that dissolution of clay material passing through the filters does not occur during the analytical treatment of samples and that only very small amounts of Al adsorbed on particles passing through the filters are detected by this technique. The) concluded that this method is a very reliable measurement of the dissolved alumintum species. Therefore. the Al,,, concentrations presented in this pi~per represent total dissolved ~Ilttrninjum or at least the total sum of reactive Al species such as those involved in chemical reactions with uluminosilicates.
To determine the effect of freezing on the analysis of AI,,, in sediment pore waters. two Xcm cores composed of homogeneous. grey. medium-graincd. quartzose sand containing calcitreous tests of molluscs (from 0 to 20 cm). and of compact. black mud (from 20 to 25 cm! were coliected m an artificial lagoon on the Belgian Coast. The cores were taken within SOcm of each other. From one of the cores. pore waters were extracted soon after collection by pressure filtration. and then stored at 5 C prior to analysis of A&, hy the fluorimetric method. The second core was sliced into S cm sections which were stored frozen in sealed polyethylene bags for 1 days. After thawing orernight at the irk sirzr temperature of collection. the pore waters were extracted from the sediments by pressure tiltrstion and stored at 5 C prior to Al,, analysis. Figure la shows the Al;,, concentrations obtained for the pore waters extracted from these frozen and unfrozen sediments. The data show that in this case freezing of sediments prior to analysis of the pore waters for Al,,, results in about a 13”,, decrease in the original Al,, concentrations. This conclusion is based on the assumption that the pore waters of the frozen and unfrozen sediments had the same original Al,, concentrations at similar core depths under irt sir~r conditions. Furthermore. the shape of the vertical concentration gradient of Al,, in pore waters from both the frozen and unfrozen sediments is very similar (Fig. la). Thus. it appears that ~ilthough sediment freezing may affect Al,,, concentrations in interstitial waters, the shape of the concentration profile of Al,, may not be significantly altered bq freezing. Also. It appears that in these experimental cores.
the absolute decrease in Al,, concentrations because of freezing is linearly related to original Al,,, concentrations (Fig. lb). Freezing of extracted mterstitmi waters however. does not seem to signi~~dntl~ a@ect Al,*, concent~tions. Pore waters were obtained h) pressure filtration from a 32 cm core collected in the Mediterranean Sea. The waters before and after freezing and thawing were analyzed for Al,,,,. The results (Table I) show that differences rn Al,,,, concentrations for waters at similar depths analyzed before and after freezmg and thawing are In the range of analytical error. In summary. although freezing of sedtments before extraction of pore waters does significantly reduce AI,, concentrations. the effect is not iarge enough to invalidate the conclusions drawn later m this paper.
RESULTS
ArVD
DISCCSSION
.As has been reported for several other dissolved speaes. the concentration of Al,,, in the mterstltinl \%atcrs of sediments IS consrderab& higher than m the overiyng water (Figs. L6t. The conc~l~tratton of Al., tn pore satrrs ma) ;rttain \aiues 3%?0 times its concentration In hottom Maters. Furthermore. the shape of the Al,,, concentration grndients in interstitial waters near the sediment-water mterface tmpltes a tlux of Al,,, from the sediments to the overlying water column. This tluv ma> be responsible for the rclativel> high Al,,,, knlues observed in deep seawater Table I. Al,,, concentration in mtersttttal waters extracted from the sediments of
I c- 10
:.02
I .a6
5
32 52
3.50
0.3
j
Notes
i
pn
Fig. 2. Vertical distribution of dissolved silica (A,). AI,,,, (I$; concentrations. and pH (ml m the interstitial waters of core 5?-2 sediments (Mediterranean Sea. Corslcdb. Duplicate analyses of Al.,,, 4.1 were achieved on stored waters (see text). (MACUKZIL. cr ul.. 1978: CASCH~T-W pubhshed). The concentration of Al,, in the Mediteranean Sea sediments offshore than m waters extracted from North
and
427
WOLLAST. to be
interstitial waters of of Corscia is greater Sea sediment cores.
FIN. 4. Vertical distribution of dissolved silrca (A). Al,,,, (01 concentrations. and pH (m) in the interstitial waters of core MI 149 sediments (North Sea. Belgian Coast ). The concentration of Al,, in pore waters of core 94 from the Mediterranean Sea (Fig. 3) shows a nearly constant value of about I pmole Al I-’ at depths greater than 2.5 cm. whereas core 57-2 (Fig. 2) from the same area exhibits distinct vertical variations in its Al,, content. This vertical variation ranges from about 0.1 to 2.6pmole Al I-’ In contrast. the concentration of Al,, in pore waters from North Sea sediments ranges from
Cm
FIN. 3. Vertical distribution of dissolved silica (A). Al,, (0) concentrations. and pH fm) in the interstitial waters of core 94 sediments (Mediterranean Sea. Corsica).
Fig. 5. Vertical distribution of dissolved silica (A) and Al,, (0) concentrations. in the interstitial waters of core I9 sedlments (North Sea. Fladden Ground).
428
Notes chemical equilibrium concentration. The stationary value results from the competition between the dissolution of biogeneous amorphous silica and uptakr of dissolved silica by silicates (WOLLAST.1974). 2. The mineral assemblage is extremely complex and several alumino-silicates may compete in the release and uptake of dissolved species. 3. As potnted out recently by LIPPMANNf 1977). description of the aqueous equilibrtum solubility of clay minerals such as illite. montmorillonite or interstratified mtermediate. requires the use of up to five partial solubtlity products. Therefore. tt is unlikely that these minerals form in nature under equilibrium conditions.
Fig. 6. Vertical distribution of dissolved silica (A) and Al,,,, (0) concentrations in the interstitial waters of core 32 sediments (North Sea. Fladden Ground).
.4ck,1on/rtlgerr1~,~tf.~-Weare grateful to the Woods Hole Oceanographic Institution and to Professor T. C. JOHMON for inviting us on the 54-6 Ktmrr crutse. The Mediterranean cores were collected on board of the Rt Rccrctrr. Ddurissor~ of the STARES0 (Calvi). We are indebted to Professor F. T. MACKENZIEfor his critical revision of the manuscript. We thank Professor J. W. MORSEfor his suggestions in reviewing this paper. Pore waters of core 57-2 were collected and measured for Al in collaboration with Mr. M. STOFF’I.\;.Mrs. N. DII~OC.G~ensured analytical support with her usual ethciency. REFERENCES
We compared the pore water results for Al,, with the aluminium concentrations obtained by HYDES(1977) during laboratory dissolution experiments of kaolinite. fireclay and montmorillonite in seawater. After four months at final pHs and H,SiO, concentrations of 7.35-8.1 and 43-207 nmole I-‘. respectively. Hydes found Al,, concentration values ranging from 0.1 to 0.3 itmole Al I“‘. In one experiment with montmorillonite at pH of 7.35. the Al,, concentration was only 0.05 pmole Al I- ‘, This range of Al,, concentrations is comparable to the range of concentrations observed in the North Sea sediment pore waters. Therefore. it is possible that Al,, in these sediments is controlled by reactions involving clay minerals. However. the A!,, concentrations in the pore waters of Mediterranean sediments are one order of magnitude higher than the experimental concentrations observed by Hydes. Ion activity products (IAP) for aluminium hydroxides were calculated for the interstitial waters and compared with equilibrium constants (K) for mineral dissolution-precipitation reactions. No consistent relationship was found to exist between IAPs and equilibrium constants. Indeed. in the case of cores Ml 149 and 94. the concentration of Al, in the interstitial waters of these sediments is S-50 times less than that required for equilibrium with aluminium hydroxides at the observed pH values. Also. there is no consistent IAP as one should expect if the concentration of the dissolved species are controlled by an equilibrium with clay minerals or other aluminosilicates. The lack of any consistent relationship between IAPs and equilibrium constants for minerals may be due to the following reasons: I. During early diagenesis dissolved silica in pore waters attains a steady-state concentration rather than a true
CASCHETTO S. and WOLLASTR. (to be published) Vertical distribution of dissolved aluminium in the Mediterranean Sea. Mar. C/IY~II. HOXXAWA 1.. OSHIMAF. and Ko~uo N. (1970) On the concentrations of the dissolved chemical elements m the estuary of the Chikugogdwa River. J. Ocrrrttoc/r. Sot. Ju(I. 26. 1-S. H~IXS J. D. (1977) Dissolved aluminium concentration m sea water. Nurtrrr 268. 136-137. HYDES J. D. and LISS P. S. (1976) Fluorimetric method for the determination of low concentrations of dissolved aluminium in natural waters. Analrsr 101, 922-931. HYDESJ. D. and LISS P. S. (1977) The behaviour of dissolved aluminium in estuarine and coastal waters. Esr. Coastal
Mar.
Sci. 5. 755-769.
LIPPMANNF. (1977) The solubility products of complex minerals. mixed crystals. and three-layer clay minerals. Nrurs Jalrrh. ,Miwrdog. Ahh. 130. 243- 263. MACKENZIEF. T.. STOFFYNM. and WOLLASTR. (1978) Aluminum in seawater: control by biological activity. Science 19% 680-682. NISHIKAWAY.. HIRAU K.. MORISHIG~K. and SHIC~MATSL T. (1967) Fluorophotometric determination of aluminum and gallium with lumogallion. Japan .4nalrsr 16, 692-697. SHIGEMATSC T.. NISHIC;AWA Y.. HtRAkt K. and NAGANO N. (1970) Fluorometric determinatton of trace amount of aluminum in natural water by lumogallion method: masking of ferric iron with 0-phenanthroline. Jupun .A!lUl!X 19, 551-555. WOLLASTR. (1974) The silica problem. In The Sea. Vol. 5 Marine Chemistr! (ed. E. D. Goldberg), pp. 359-392. Wiley.
Dissolved
aluminium in interstitial marine sediments SKRG~ CASCH~:TKJ and
Laboratoire
d’Oc&nographie.
Universitti
Librc
waters of recent
ROLANI) WOLLAST
de Bruxelles.
A\. F. Roosevelt.
50. Bruxelles
1050. Belgrum
Abstract-Measurements of the concentration and vertical distrlhution of dissolved aluminium and silica. and of pH. in interstitial waters of recent marine sediments from the North Sea and the Meditcrranean Sea were performed to evaluate the behaviour of aluminium during early diagenesis. The results suggest that thermodynamic equilibria alone do not control the concentratrons of dissolved species in the system AI-Si-O,-H,O during early diagenesis. Rather. these concentrations are governed by dynamic factors involving mineral dissolution-precipitation reactions and diffusion.
INTRODUCTION Knowledge of the concentration and distribution of dissolved aluminium (Al,,,,) in natural waters has improved substantialI> since the development of Al fluorimetric method. In the case of seawater. the vertical distribution of Al,, in the water column strongly suggests the importance of the planktonic activity on the removal of AI from the oceanic system (MA~KI:NZIL CI al.. 1978; CASCHETTO and WOLLAST. to be published). The behaviour of Al, in estuarine environments is still a matter of controversy (HBOKAWA (‘I ul.. 1970: HYUES and LISS. 1977). To our knowledge. there are no data available on the distribution of Al,., in the interstitial waters of marine sedlmen&. However. to understand the geochemical cycle of Al and processes involved in the earl) diagenesis of cla! minerals. it is necessarv to know the concentration. distribution. and mineral sinks and sources of Al,,., in marine sediments. This brief contribution provides some preltmmarq results from coastal marine sediments concerning these factors. METHODOLOCI Gene-al
Five cores were collected. two from the Mediterranean Sea and three from the North Sea. The Mediterranean cores 57-2 and 94 were obtained using a gravity corer about 2.7 km offshore of Calvi (Corsica) at depths of 57 and 94 m. The core sediments consist principally of medium-grained. quartzo-feldspathic sand containing accessor! biotite. calcareous tests. siliceous sponge spicules and numerous remains of marine phanerogams. Cores 19 and 32 were taken in the Fladden Ground of the North Sea (58.25’5”N. 00’02’O”E and 58’50’3”N. 00’41’4”E) at depths of 135 and 140m by means of a box-corer during cruise 54-6 of the K~~orr (“FLEX” program). These sediments are medium-grained sands. Glacial pebbles are recognizable in core 33. The third North Sea core was collected b) Belgian Nav) divers about 1.8 km offshore of the Belgian Coast at a depth of 10m (Station Mll49). It IS characterized b> a ver) fine. dark mudd! sand. No laminations are present in the five cores.
Core 57-2 was shced Immcdlatel~ after collection on board ship. and then stored in sealed polyethylene bags for 3 hr at IO C. The interstitial waters were extracted from the sediments using a PVC ‘squeezer’ by pressure filtration through a 0.45 blrn Milhpore” filter under 3 bar of nitrogen. Cores 19. 32.94. and M 1 149 were frozen immediately after collection. The interstitial waters of cores 19. 32 and 94 were later extracted following the above procedure. Waters from core Ml 149 however. were removed from the sediments by centrifugation and subsequent filtration. In all cases. the pore waters were extracted a~ the i/l .sirrr temperatures of collection. Frozen cores were thawed overnight at the iu .\itu temperature before treatment.
Interstitial waters were analyzed for dissolved silica b) absorption spectrophotometrq (Technicon Auto-Analyzer. Methodology 186-72W) and for dissolved aluminium by fluorescence spectrophotometry of the Al-lumogallion complex (NISHIKAWA (0 ul.. 1967: SHIGKMATSI. or al.. 1970: HYI>ES and LISS. 1976). using the method of Al increments. In all the pore waters we analyzed. this Al incremental calibration was able to prevent the problems involving interferences owing to F. SOi-. Fe and dissolved organic matter as previously shown by Hsuts and Ltss (1976). However. if sediments contain much organic matter. it must be photo-oxidized (IOOOW U.V. Hanovia lamp) prior to the AI,, analysts. as was done in the case of core M 1 149. The practical detectlon limit of the Al-lumogallion method was found to be 0.003~mole Al I-‘. From five independent analyses of a seawater containing 0.078 pmole Al I-‘, we obtained an estimated standard deviation of 0.003 pmole Al I_ ‘. To determine the effect of storage on Al,, determinations. a duplicate analysis of Al,, was performed on the interstitial waters of core 57-2 aker storage of the waters for IO days at 7 C (Fig. 2). After storage most of the water samples exhibited a decrease in Al,, but the general shape of the vertical concentration gradient of Al,, was maintained. HYDE and LISS 11976) investigated variations m the amount of AI,,, detected bj this fluorimetric method in claq suspensions of various concentrations and in natural waters which were filtered through different pore size mem425
Nom
426
OcDth
I
unfrozen
a
sediment
Al,,
( PMI-’ I
b
Fig. I. Al.,, concentration in interstitial waters of sediments collected m ,tn artifiaal lagoon on the Belgian Const: (al comp~t~son between the vertical distribLition in mterstittal waters squeezed from the unfroam sediment (0) and the vertical distributton in interstitial waters squeezed from frozen sediment ( + I: (b) plot of the values from unfrozen sediment tersus those from frozen sediment. The slope of the line through the points IS 0.X7. hranrs. They were able to show that dissolution of clay material passing through the filters does not occur during the analytical treatment of samples and that only very small amounts of Al adsorbed on particles passing through the filters are detected by this technique. The) concluded that this method is a very reliable measurement of the dissolved alumintum species. Therefore. the Al,,, concentrations presented in this pi~per represent total dissolved ~Ilttrninjum or at least the total sum of reactive Al species such as those involved in chemical reactions with uluminosilicates.
To determine the effect of freezing on the analysis of AI,,, in sediment pore waters. two Xcm cores composed of homogeneous. grey. medium-graincd. quartzose sand containing calcitreous tests of molluscs (from 0 to 20 cm). and of compact. black mud (from 20 to 25 cm! were coliected m an artificial lagoon on the Belgian Coast. The cores were taken within SOcm of each other. From one of the cores. pore waters were extracted soon after collection by pressure filtration. and then stored at 5 C prior to analysis of A&, hy the fluorimetric method. The second core was sliced into S cm sections which were stored frozen in sealed polyethylene bags for 1 days. After thawing orernight at the irk sirzr temperature of collection. the pore waters were extracted from the sediments by pressure tiltrstion and stored at 5 C prior to Al,, analysis. Figure la shows the Al;,, concentrations obtained for the pore waters extracted from these frozen and unfrozen sediments. The data show that in this case freezing of sediments prior to analysis of the pore waters for Al,,, results in about a 13”,, decrease in the original Al,, concentrations. This conclusion is based on the assumption that the pore waters of the frozen and unfrozen sediments had the same original Al,, concentrations at similar core depths under irt sir~r conditions. Furthermore. the shape of the vertical concentration gradient of Al,, in pore waters from both the frozen and unfrozen sediments is very similar (Fig. la). Thus. it appears that ~ilthough sediment freezing may affect Al,,, concentrations in interstitial waters, the shape of the concentration profile of Al,, may not be significantly altered bq freezing. Also. It appears that in these experimental cores.
the absolute decrease in Al,, concentrations because of freezing is linearly related to original Al,,, concentrations (Fig. lb). Freezing of extracted mterstitmi waters however. does not seem to signi~~dntl~ a@ect Al,*, concent~tions. Pore waters were obtained h) pressure filtration from a 32 cm core collected in the Mediterranean Sea. The waters before and after freezing and thawing were analyzed for Al,,,,. The results (Table I) show that differences rn Al,,,, concentrations for waters at similar depths analyzed before and after freezmg and thawing are In the range of analytical error. In summary. although freezing of sedtments before extraction of pore waters does significantly reduce AI,, concentrations. the effect is not iarge enough to invalidate the conclusions drawn later m this paper.
RESULTS
ArVD
DISCCSSION
.As has been reported for several other dissolved speaes. the concentration of Al,,, in the mterstltinl \%atcrs of sediments IS consrderab& higher than m the overiyng water (Figs. L6t. The conc~l~tratton of Al., tn pore satrrs ma) ;rttain \aiues 3%?0 times its concentration In hottom Maters. Furthermore. the shape of the Al,,, concentration grndients in interstitial waters near the sediment-water mterface tmpltes a tlux of Al,,, from the sediments to the overlying water column. This tluv ma> be responsible for the rclativel> high Al,,,, knlues observed in deep seawater Table I. Al,,, concentration in mtersttttal waters extracted from the sediments of
I c- 10
:.02
I .a6
5
32 52
3.50
0.3
j
Notes
i
pn
Fig. 2. Vertical distribution of dissolved silica (A,). AI,,,, (I$; concentrations. and pH (ml m the interstitial waters of core 5?-2 sediments (Mediterranean Sea. Corslcdb. Duplicate analyses of Al.,,, 4.1 were achieved on stored waters (see text). (MACUKZIL. cr ul.. 1978: CASCH~T-W pubhshed). The concentration of Al,, in the Mediteranean Sea sediments offshore than m waters extracted from North
and
427
WOLLAST. to be
interstitial waters of of Corscia is greater Sea sediment cores.
FIN. 4. Vertical distribution of dissolved silrca (A). Al,,,, (01 concentrations. and pH (m) in the interstitial waters of core MI 149 sediments (North Sea. Belgian Coast ). The concentration of Al,, in pore waters of core 94 from the Mediterranean Sea (Fig. 3) shows a nearly constant value of about I pmole Al I-’ at depths greater than 2.5 cm. whereas core 57-2 (Fig. 2) from the same area exhibits distinct vertical variations in its Al,, content. This vertical variation ranges from about 0.1 to 2.6pmole Al I-’ In contrast. the concentration of Al,, in pore waters from North Sea sediments ranges from
Cm
FIN. 3. Vertical distribution of dissolved silica (A). Al,, (0) concentrations. and pH fm) in the interstitial waters of core 94 sediments (Mediterranean Sea. Corsica).
Fig. 5. Vertical distribution of dissolved silica (A) and Al,, (0) concentrations. in the interstitial waters of core I9 sedlments (North Sea. Fladden Ground).
428
Notes chemical equilibrium concentration. The stationary value results from the competition between the dissolution of biogeneous amorphous silica and uptakr of dissolved silica by silicates (WOLLAST.1974). 2. The mineral assemblage is extremely complex and several alumino-silicates may compete in the release and uptake of dissolved species. 3. As potnted out recently by LIPPMANNf 1977). description of the aqueous equilibrtum solubility of clay minerals such as illite. montmorillonite or interstratified mtermediate. requires the use of up to five partial solubtlity products. Therefore. tt is unlikely that these minerals form in nature under equilibrium conditions.
Fig. 6. Vertical distribution of dissolved silica (A) and Al,,,, (0) concentrations in the interstitial waters of core 32 sediments (North Sea. Fladden Ground).
.4ck,1on/rtlgerr1~,~tf.~-Weare grateful to the Woods Hole Oceanographic Institution and to Professor T. C. JOHMON for inviting us on the 54-6 Ktmrr crutse. The Mediterranean cores were collected on board of the Rt Rccrctrr. Ddurissor~ of the STARES0 (Calvi). We are indebted to Professor F. T. MACKENZIEfor his critical revision of the manuscript. We thank Professor J. W. MORSEfor his suggestions in reviewing this paper. Pore waters of core 57-2 were collected and measured for Al in collaboration with Mr. M. STOFF’I.\;.Mrs. N. DII~OC.G~ensured analytical support with her usual ethciency. REFERENCES
We compared the pore water results for Al,, with the aluminium concentrations obtained by HYDES(1977) during laboratory dissolution experiments of kaolinite. fireclay and montmorillonite in seawater. After four months at final pHs and H,SiO, concentrations of 7.35-8.1 and 43-207 nmole I-‘. respectively. Hydes found Al,, concentration values ranging from 0.1 to 0.3 itmole Al I“‘. In one experiment with montmorillonite at pH of 7.35. the Al,, concentration was only 0.05 pmole Al I- ‘, This range of Al,, concentrations is comparable to the range of concentrations observed in the North Sea sediment pore waters. Therefore. it is possible that Al,, in these sediments is controlled by reactions involving clay minerals. However. the A!,, concentrations in the pore waters of Mediterranean sediments are one order of magnitude higher than the experimental concentrations observed by Hydes. Ion activity products (IAP) for aluminium hydroxides were calculated for the interstitial waters and compared with equilibrium constants (K) for mineral dissolution-precipitation reactions. No consistent relationship was found to exist between IAPs and equilibrium constants. Indeed. in the case of cores Ml 149 and 94. the concentration of Al, in the interstitial waters of these sediments is S-50 times less than that required for equilibrium with aluminium hydroxides at the observed pH values. Also. there is no consistent IAP as one should expect if the concentration of the dissolved species are controlled by an equilibrium with clay minerals or other aluminosilicates. The lack of any consistent relationship between IAPs and equilibrium constants for minerals may be due to the following reasons: I. During early diagenesis dissolved silica in pore waters attains a steady-state concentration rather than a true
CASCHETTO S. and WOLLASTR. (to be published) Vertical distribution of dissolved aluminium in the Mediterranean Sea. Mar. C/IY~II. HOXXAWA 1.. OSHIMAF. and Ko~uo N. (1970) On the concentrations of the dissolved chemical elements m the estuary of the Chikugogdwa River. J. Ocrrrttoc/r. Sot. Ju(I. 26. 1-S. H~IXS J. D. (1977) Dissolved aluminium concentration m sea water. Nurtrrr 268. 136-137. HYDES J. D. and LISS P. S. (1976) Fluorimetric method for the determination of low concentrations of dissolved aluminium in natural waters. Analrsr 101, 922-931. HYDESJ. D. and LISS P. S. (1977) The behaviour of dissolved aluminium in estuarine and coastal waters. Esr. Coastal
Mar.
Sci. 5. 755-769.
LIPPMANNF. (1977) The solubility products of complex minerals. mixed crystals. and three-layer clay minerals. Nrurs Jalrrh. ,Miwrdog. Ahh. 130. 243- 263. MACKENZIEF. T.. STOFFYNM. and WOLLASTR. (1978) Aluminum in seawater: control by biological activity. Science 19% 680-682. NISHIKAWAY.. HIRAU K.. MORISHIG~K. and SHIC~MATSL T. (1967) Fluorophotometric determination of aluminum and gallium with lumogallion. Japan .4nalrsr 16, 692-697. SHIGEMATSC T.. NISHIC;AWA Y.. HtRAkt K. and NAGANO N. (1970) Fluorometric determinatton of trace amount of aluminum in natural water by lumogallion method: masking of ferric iron with 0-phenanthroline. Jupun .A!lUl!X 19, 551-555. WOLLASTR. (1974) The silica problem. In The Sea. Vol. 5 Marine Chemistr! (ed. E. D. Goldberg), pp. 359-392. Wiley.